The application of the Internet of Things involves all aspects, in industry, agriculture, environment, transportation, logistics, security and other infrastructure areas, effectively promoting the intelligent development of these areas, making the more rational use of limited resources allocation, thereby improving the efficiency and effectiveness of the industry.
In the home, medical and health, education, finance and services, tourism and other closely related to life in the field of application, from the scope of services, services to the quality of services and other aspects have been greatly improved, greatly improving the quality of people's lives;
Involved in the military field of national defense, although it is still in the stage of research and exploration, but the application of the Internet of things brought about by the impact is also Can not be underestimated, large to satellite, missiles, aircraft, submarines and other equipment systems, small to single combat equipment, the Internet of Things technology embedded in the effective enhancement of military intelligence, information technology, precision, greatly enhance the military combat effectiveness, is the key to the future of military change.
I. Intelligent Transportation
The application of Internet of Things technology in road traffic is relatively mature. With the increasing popularity of social vehicles, traffic congestion and even paralysis has become a major problem in the city. Real-time monitoring of road traffic conditions and the timely transmission of information to the driver, allowing drivers to make timely travel adjustments, effectively alleviating traffic pressure;
Highway intersections to set up road automatic toll collection system (referred to as ETC), eliminating the need for import and export to pick up the card, the time to return the card to enhance the efficiency of the vehicle's passage; buses installed on the positioning system, which can be timely to understand the bus routes and arrival time, and passengers can determine the route and arrival time of buses. Passengers can determine the travel according to the ride route, eliminating unnecessary waste of time.
The increase of social vehicles, in addition to bringing traffic pressure, parking difficulties are increasingly becoming a prominent issue, many cities have launched a smart street parking management system, which is based on a cloud computing platform, combined with Internet of Things technology and mobile payment technology, *** enjoy parking resources, improve the utilization rate of the parking space and the user's convenience.
The system is compatible with cell phone mode and radio frequency identification mode, through the cell phone APP software can realize timely understanding of the parking space information, the location of the parking space, make a reservation in advance and realize the payment of fees, etc., to a large extent, to solve the problem of "difficult to park, difficult to park".
Two, smart home
Smart home is the Internet of Things in the family's basic applications, with the popularization of broadband services, smart home products involved in all aspects. No one at home, you can use cell phones and other product clients to remotely operate intelligent air conditioning, adjust the room temperature, and even learn the user's habits, so as to achieve fully automatic temperature control operation, so that the user in the hot summer to go home to enjoy the comfort of ice;
Through the client to achieve the smart bulb on and off, regulating the brightness of the bulb and color, etc.; socket built-in Wifi, can be achieved by remote control! The socket built-in Wifi, can realize the remote control of the socket timer on and off current, even can monitor the equipment power consumption, generate power consumption charts so that you can get a clear picture of the power situation, to arrange the use of resources and spending budget;
Smart scale, monitor the effect of exercise. Built-in advanced sensors that can monitor blood pressure, fat volume, built-in program to make health recommendations based on the state of the body; Smart toothbrush connected to the client for brushing time, brushing position reminder, can be based on brushing data production charts, the health of the oral cavity;?
Smart cameras, window sensors, smart doorbells, smoke detectors, smart alarms, etc. are indispensable home security monitoring equipment, you go out in time to view the real-time status of any corner of the home at any time, place, any security risks. Seemingly cumbersome all kinds of home life because of the Internet of Things has become easier and better.
Three, public **** security
In recent years, the global climate anomalies are frequent, the suddenness of the disaster and the hazards of further increase, the Internet can real-time monitoring of the environment's insecurity, prevention, real-time early warning, and timely response to reduce the threat of disasters to human life and property.?
The University of Buffalo in the United States as early as 2013 proposed a research project on deep-sea Internet, through a special treatment of the induction device placed in the deep sea, analyzing the relevant underwater situation, the prevention and control of marine pollution, the detection of undersea resources, and even tsunamis can provide a more reliable early warning. The project was successfully tested in local lake water, providing a basis for further expansion of its use.
Using IoT technology can intelligently sense data on various indicators in the atmosphere, soil, forests and water resources, playing a huge role in improving the human living environment.
Trends and Characteristics
The main notable trend in IoT in recent years has been the explosion of devices connected and controlled by the Internet. The widespread adoption of IoT technology means that the specifics of moving from one device to another can vary widely, but most share basic characteristics.
The IoT creates the opportunity to integrate the physical world more directly into computer-based systems, resulting in increased efficiency, economic benefits and reduced labor.
The number of IoT devices grew 31% year-on-year to 8.4 billion in 2017 and is expected to be 30 billion by 2020. The global market value of IoT is expected to reach $7.1 trillion by 2020.
Environmental intelligence and autonomous control were not part of the original concept of the IoT. Nor do ambient intelligence and autonomous control necessarily require an Internet structure. However, there has been a shift in research (by Intel and others) to combine the concepts of IoT and autonomous control, and initial results are moving in that direction, looking at objects as drivers of the autonomous IoT.
One promising approach in this context is deep reinforcement learning, where most IoT systems provide dynamic and interactive environments. Training agents (i.e., IoT devices) to perform smarter in such environments cannot be addressed by traditional machine learning algorithms such as supervised learning.
With a reinforcement learning approach, a learning agent can sense the state of the environment (e.g., sense the temperature of the home), perform an action (e.g., turn the HVAC on or off) and learn by maximizing the cumulative rewards it receives over time.
IoT intelligence can be provided at three levels: the IoT device, the edge/fog node and the cloud. The need for intelligent control and decision-making at each level depends on the time-sensitivity of the IoT application. For example, a self-driving car's camera needs to perform real-time obstacle detection to avoid an accident.
This kind of rapid decision-making is not possible by transferring data from the vehicle to a cloud instance and returning predictions to the vehicle. Instead, all operations should be performed locally in the vehicle. Integrating advanced machine learning algorithms, including deep learning IoT devices is an active area of research to bring smart objects closer to reality.
In addition, maximum value can be gained from IoT deployments by analyzing IoT data, extracting hidden information, and predicting control decisions. A wide variety of machine learning techniques are used in the IoT field, ranging from traditional methods such as regression, support vector machines and random forests to advanced methods such as convolutional neural networks, LSTMs and variational autoencoders.
In the future, the IoT is likely to be a non-deterministic and open network in which automatically organized or intelligent entities (Web services, SOA components) and virtual objects (avatars) will be interoperable and capable of acting independently (pursuing their own goals) goals or *** enjoyment of goals) depending on the context, situation or environment.
Autonomous behavior through the collection and reasoning of contextual information and the ability of objects to detect changes in the environment (malfunctions affecting sensors) and to introduce appropriate mitigations constitutes a major research trend that is clearly needed to provide credibility to IoT technologies.
Modern IoT products and solutions on the market use a variety of different technologies to support this context-aware automation, but more sophisticated forms of intelligence are needed to allow the deployment of sensor units and intelligent cyber-physical systems in real environments.
The above reference? Baidu Encyclopedia - Internet of Things